Browse Prior Art Database

An intelligent semantics check and alert system in social communication Disclosure Number: IPCOM000248924D
Publication Date: 2017-Jan-22
Document File: 6 page(s) / 167K

Publishing Venue

The Prior Art Database


You are what you say, that is, you build your personal branding with those words you said to others. This is especially the case with your personal image on social media. However, though the market has already provided users with tools, such as Spelling Checker, to help users get rid of unexpected typos, other inappropriate messages which are caused by indecent words or multiple-meanings emoji are still beyond existing solutions’ hand. As a result, those error messages lead to confusion, misunderstanding, and even awkward moments, which in a way or another interrupt your conversation and do harm to the communication efficiency. Therefore, we propose an intelligent semantics check and alert system to help users to notice and modify words, or in some cases, emojis, which are correct-in-spelling but inappropriate-under-scenario, before they send these messages out.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 52% of the total text.


An intelligent semantics check and alert system in social communication

Part 1: Pain point illustration

As it has been talked in the abstract part, this invention is aiming to get rid of inappropriate messages under the instant messaging scenario. An example of the pain point scenario has been illustrated, see Fig 1 below.

(Fig 1. An illustration of a typical pain point scenario )

Part 2: Description of this invention

This systems offers you options with which you can customize this filtering system beforehand based on your relationship with each message receiver. You can define restrict or loose filtering rules as you prefer. Meanwhile, it collects and analyze those social data, such as previous instant message records, commons you both share, etc., to offer you with a recommended rule setting to help you improve social communication effects.


To save users’ efforts to manually shift the restricted words/emoji list contact by contact, here, we bring up a conception namedatmosphere level to help the communication system intelligently achieve a dynamic level-based list which shifts by contact, in the backstage. So, you might will ask: "What is the atmosphere level?". Here, we define the atmosphere level as a measure/ an indicator to show how close/warm the relation between the message sender and the message receiver. It could be an ordinal variable, such like level I, level II and level III; or just simply be an exact number out of 10 or 100 or any other number to illustrate the specific degree level.

2.1 How it works Generally, this semantics check and alert system works with the following logic, see fig 2 below.


(Fig 2. How does this alert system works)

2.2 Usage Scenario

Step 1: A user add a new contact with the “Semantic-check mode” activated. Since there is no chat history data which the communication system could use to analyze and then rank the atmosphere level accordingly, so the system will call up all related data to help assign an accurate-enough starting atmosphere level. e.g. If a new contact comes from an existing chat group which the user joined, or has been sort into a specific relation group (such as families, colleagues, friends, report chain), then this new contact’s starting atmosphere level could be calculated as the mean level of this group. See figure 3. If a new contact is...